The IP address in the TCP/IP (Transmission Control Protocol/Internet Protocol) widely used in the existing Internet has a dual function, and it can act as the locator of the correspondent terminal host network interface of the network layer in the network topology and also the identification of the host network interface of the transport layer. At the first stage of the TCP/IP protocol design, the moving of the host is not considered, however, when the moving of the host is more and more common, the semantic overload defect of this IP address has become increasingly evident. When the IP address of a host changes, not only the route but also the identification of the communication terminal host change, which will lead to heavier and heavier routing load, and moreover, the change of the host identifier will interrupt the connections and the applications. The purpose of proposing the identification and location separation problem is to solve problems such as the semantic overload of the IP address and the too heavy routing load, to separate the dual function of the IP address so as to achieve supporting to problems such as mobility, multi-home, dynamic IP address re-allocation, routing load reduction and the exchange visits between different network regions in the next generation Internet.
In the related art, the network router based method is one of the solutions to separate the identification and the location. In the related art, a method for implementing the mobile switching management based on the network router solution is proposed, in this method, the access switch router notifies the access switch router in which the correspondent nodes (referred to as CN) of all the mobile terminals are located of the new mapping information of a mobile node (MN).
In the switch process in the related art, the information of all the CNs is saved in the access switch router for each MN that accesses locally, as shown in the following table:
IdentificationLocatorCN 1. . .CN N
In the switch process, the access switch router at the switch-out side needs to send this correspondent node table to the access switch router at the switch-in side, the access switch router at the switch-in side notifies the access switch router in which the CN is located of the MN's new identifier mapping information based on this information.
There are mainly the following problems in the switch process in the related art:
1. for many service applications, such as the P2P (peer-to-peer) and so on, in the existing network, each MN has dozens or even hundreds of CNs and it is needed to establish, store and maintain the correspondent node table in network devices such as the access switch router, the access service node (referred to AN), or the identity location register, which needs to take up a lot of memory and CPU processing performance and needs complicated signaling flow processing between the devices;
2. when data communication is performed between the MN and the CN, the AN at the switch-out side of the MN needs to synchronize information of all the CNs of the MN to the identity location register (referred to as ILR), the ILR stores the information of all the CNs communicating with the MN and maintains the accuracy of the information of the correspondent node through signaling messages between the ILR and the AN. In the switch process, the AN at the switch-in side allocates a new RID to the MN, and a mapping update process is initiated between the ILR and the AN at the switch-in side or the switch-out side. After the mapping update flow ends, the ILR notifies the AN where all the CNs are located of the MN's new identifier mapping information according to the saved information of the CNs;
alternatively, in the switch process, the AN at the switch-out side synchronizes the information of all the CNs of the MN to the AN at the switch-in side, the AN at the switch-in side saves the information of all the CNs of the MN, after a new RID is allocated to the MN, the new mapping information of the MN is notified to the AN where all the CNs are located according to the saved information of all the CNs;
3. after the data communication relationship is established between the MN and the CN, the AN in which the CN is located saves the real-time identification-locator mapping information of the MN. The purpose of the network device establishing, storing and maintaining the correspondent node table is to guarantee that the correspondent node can be notified timely in the case that the accessed MN is allocated with new routing identifier, so as to ensure that the identification-locator mapping information of the correspondent node saved by the AN is real-time effective, thereby the data message sent by the CN are able to be routed to the AN at the switch-in side of the MN after the newest RID is encapsulated in the data message. After the peer-to-peer connection between the MN and the CN is disconnected, the AN in which the MN is located no longer notifies the AN in which the CN is located of the mapping information change of the MN. Therefore, after the peer-to-peer connection between the MN and the CN is disconnected, the network device needs to delete the MN from the CN's correspondent node table. When all terminals accessing to the AN in which the CN is located have no communication with the MN, the mapping and routing table entry of the MN should be deleted from the AN in which the CN is located. Wherein, when the network device needs to delete the MN from the CN's correspondent node table, the AN in which the CN is located needs to identify and monitor the peer-to-peer connection which is identified by the MN-CN, and determine the connected/disconnected state of the peer-to-peer connection, which requires that the AN in which the CN is located forwards all the data messages in the way of data streams, if a certain data stream has no data to send in a time period, the data stream is considered as in the disconnected state. Compared with the method used in the existing routers and other equipments for forwarding the data messages one by one, forwarding data in the way of data streams increases the load for processing data stream establishment, data stream mapping, and data stream connected/disconnected monitoring and so on, which makes the implementation of AN complicated.